Does Participatory Budgeting Bolster Voter Turnout in Elections ? The Case of the Czech Republic - Sciendo

Page created by Clinton Hampton
 
CONTINUE READING
The NISPAcee Journal of Public Administration and Policy, Vol. XII, No. 2, Winter 2019 /2020

10.2478/nispa-2019-0016

Does Participatory Budgeting Bolster Voter Turnout
in Elections ? The Case of the Czech Republic
                                                                 Soňa Kukučková1, Eduard Bakoš2

Abstract
Though participatory budgeting (PB) is often discussed as a tool to bolster the level
of civic participation and the quality of democracy, empirical research on the sub-
ject offers ambiguous results. In the Czech Republic, PB was introduced 5 years ago,
and the number of implemented PBs has since increased substantially. The purpose
of this article is to evaluate whether the use of PB is associated with higher voter
turnout in municipal and parliamentary elections. Voter turnout in Czech munic-
ipalities that implemented PB is analyzed and compared with the control group of
municipalities without PB. Considered by type of election, we found that the impact
of PB use on voter turnout is higher for local elections than it is for national elec-
tions, which is in line with our assumptions. However, our results were significant
for Prague districts only. Participatory budgeting could increase voter turnout in
local election, but there are other factors that must be considered.

Keywords:
participatory budgeting, voter turnout, direct democracy

Introduction
In recent years, participatory budgeting (PB) has often been discussed as a tool to
bolster the level of civic participation and the quality of democracy (e.g. Matovu
and Mumvuma 2008; Wampler 2012). In this context, PB is seen as a form of citizen
initiative which is similar to the instruments of direct democracy.

1    AssiDepartment of Finance, Faculty of Business and Economics, Mendel University in Brno,
     Czech Republic.
2    Department of Public Economics, Faculty of Economics and Administration, Masaryk University,
     Brno, Czech Republic.

                                                                                                           109
    Open Access. © 2019 Soňa Kukučková, Eduard Bakoš, published by Sciendo.

            This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.
The NISPAcee Journal of Public Administration and Policy, Vol. XII, No. 2, Winter 2019 /2020

      Tolbert et al. (2001) state that the process of direct democracy itself could
influence the attitudes and behaviors of individuals and motivate them to be more
involved in the decision-making process about public goods and services. One of
the educative effects of direct democracy could be the positive influence that direct
democracy has on voter turnout. Similarly, in Sintomer et al. (2013) PB has also
often been associated with the mobilization of citizens.
      In the EU, the European elections tackle the problem of a low voting turnout.
Since 1979 the participation rate has fallen by 20 percentage points to 42.61 %. In
the Czech Republic, the voting turnout in the European election in 2014 was the
second lowest of all EU countries with a 10-percentage point decrease compared to
the previous election turnout in 2009 to 18.2 % (European Parliament 2014). The
situation has changed slightly this year (2019) when the overall turnout in the Eu-
ropean election increased to 50.95 %. However, although voter turnout in the Czech
Republic has increased to 28.72 %, the Czech Republic remains the country with
the second lowest participation (European Parliament 2019). The possibility to mo-
tivate Czech citizens to vote and to increase their civic engagement and interest in
politics through the implementation of participatory budgeting could be presented
as a solution to the “crisis of democracy.” However, the type of election could have a
different impact on voter turnout (e.g. Tolbert et al. 2001). For example, in the local
elections in the Czech Republic, the situation is better than in the European elec-
tions (for citizens it is much more interesting at the local level because they are more
concerned about it), but in the last two elections (2014 and 2018) the turnout did
not even reach 50 %. Therefore, the article is focused on municipal elections because
we anticipate PB to have the most impact on the relationship between citizens and
municipal officials. We will also test the possible impact of PB implementation on
the parliamentary voter turnout at the national level.
      The contribution of the article can be seen in two aspects. Firstly, the objective
of the paper is to evaluate whether the use of participatory budgets is associated
with a higher voter turnout in parliamentary and municipal elections. The empir-
ical studies for European countries do not provide unambiguous results referring
to the relationship of PB implementation and voter turnout (e.g. Sintomer et al.
2014), and less attention is paid to the probable difference in the case of local and
parliamentary elections.
      Secondly, participatory budgeting is a new phenomenon in the Czech Repub-
lic. The first PB was introduced 5 years ago, and the number of PBs was growing
relatively rapidly in the period 2016 – 2018. So far, we have not found a publicly
available database or a systematic report about participatory budgeting including
all Czech PBs, only some partial analyses were published (e.g. Haken et al. 2016).
Therefore, we had to use public web-based information to complete the data, and
our preliminary results are based on a relatively lower number of finished Czech
PB cases.

110
Does Participatory Budgeting Bolster Voter Turnout in Elections ? The Case of the…

1. Participatory democracy and the objectives of participatory
   budgeting
Participatory budgeting could be understood as a combination of elements of direct
or semi-direct democracy with representative democracy (UN-Habitat 2004). In
participatory democracy, the citizen’s participation is not limited to the act of voting
to elect their own representatives and an executive but persists during and between
election periods. Participatory budgeting (PB) is one of the modern global trends of
involving citizens in deciding where and how to divide a part of a public budget. It is
a decision-making process through which citizens deliberate and negotiate over the
distribution of public resources (Wampler 2007). Sintomer et al. (2008, 2014) define
5 criteria necessary for PB to meet the definition of a proper PB process that have
been accepted as a standard around the world: financial and / or budgetary aspects
must be discussed; the city level or a decentralized district with an elected body
and some power over administration and resources has to be involved; it has to be
a repeated process over years; some forms of public deliberation must be included
within the framework of specific meetings / forums; some accountability on the re-
sults of the process is required.
      The first participatory budgets in line with Sintomer et al.’s (2008, 2014) defi-
nition were implemented in Brazil, and are named the Porto Alegre type after the
municipality where the success of the PB served as an inspiration for spreading the
practice worldwide. PBs were implemented in Europe with some modification, and
therefore the term “Porto Alegre adopted in Europe” is used (e.g. Džinić et al. 2016
or Boc, 2019). In our article we analyze the project oriented PBs of the Porto Alegre
in Europe type that prevail in the Czech Republic.
       PBs were implemented in Europe with some modification, and since then
more types of PB have been distinguished (e.g. Cabannes 2004; Sintomer et al.
2008). The type of PB described as Porto Alegre adopted for Europe is character-
ized in Džinić et al. (2016, 34) as: “discussion in neighborhood and / or thematic
assemblies primarily dealing with concrete investment and projects and their prior-
itization.” In our article we analyze the project PBs of the Porto Alegre adapted for
Europe type that prevail in the Czech Republic. The form of civic participation is di-
rect – a citizen can vote on a project as an individual without delegates. This feature
slightly differs from the definition of Porto Alegre adapted for Europe in Krenjova
and Raudla (2013), where the council as the decision-making body is composed
of citizens’ elected delegates. Our definition is in line with Gondášová and Murray
Svidroňová (2016) using this type for project PBs in Slovakia.
      Carroll et al. (2016) summarized reasons why PB is implemented and defined
six angles attracting the most interest: democracy (D), transparency (T), education
(Ed), efficiency (Ef), social justice (SJ) and community (C). Democracy means that
ordinary people have decision-making power (e.g. Sintomer et al. 2013). This process

                                                                                   111
The NISPAcee Journal of Public Administration and Policy, Vol. XII, No. 2, Winter 2019 /2020

can increase trust in government (Berman 1997) and the will of inhabitants to partic-
ipate in public activities and to vote (e.g. Matovu and Mumvuma 2008; UN-Habitat
2004). Thus, PB could be defined as a tool to improve democratic governance.
      Jacobi (1999) stresses transparency as one of the main features of the PB pro-
cess. When community members decide on spending through a public vote, there
are less opportunities for corruption, waste, or backlash (e.g. UN-Habitat 2004; Ma-
tovu and Mumvuma 2008; Wampler 2007, Carroll et al. 2016).
      Through education, participants become more active and informed citizens as
participatory-budgeting programs can serve as “citizenship schools” that allow par-
ticipants to learn democracy by doing it (Wampler 2007). Engaged citizens can gain
a better understanding of general political issues and community needs (Carroll et
al. 2016; Baiocchi and Ganuza 2014).
     Viewed from the efficiency angle, budget decisions are improved when they
draw on residents’ local knowledge and oversight. Once they are invested in the
process, people can apply pressure to make sure that money is spent wisely (Sin-
tomer et al. 2010; UN-Habitat 2004).
     From a social-justice perspective, participatory budgeting ensures that every-
one has equal access to a public decision-making process. Traditionally underrepre-
sented groups (e.g. seniors) often participate more than usual in PB, and it provides
opportunities to solve poverty problems and to mitigate social exclusion (e.g. Gold-
frank 2007; Wampler 2007, 2012; Bhatnagar et al. 2003).

                                     Figure 1
                 Understanding six angles of participatory budgeting

                                      Democracy

                  Community                                 Transparency

                                      ParƟcipatory
                                       budgeƟng

                    Social
                                                           Education
                    Justice

                                       Efficiency

                 Source: own elaboration based on Caroll et al. 2016

112
Does Participatory Budgeting Bolster Voter Turnout in Elections ? The Case of the…

      Finally, the community perspective reflects that, through the regular meetings
involved with PB, people can get to know their neighbors and feel connected to
their city. Budget assemblies connect community groups and activate them (Mat-
ovu and Mumvuma 2008).
      Similarly, Goldfrank (2007) summarized results of PB in the analogous areas:
redirecting public resources towards poor neighborhoods (akin to SJ), extending
service provision, democratizing existing and spurring the creation of new civic
associations (D and Ed), increasing transparency (T) and accountability (similar to
Ef) and enhancing democratic representation for the formerly excluded (D).
       The reasons for PB implementation have evolved since their initiation in
South America through their current application in European countries. The initial
idea of PBs in Brazil was concentrated on social justice issues (SJ), with the aim to
solve extreme disparities in income and quality of life between the rich and poor
(Sintomer et al. 2010, 2014; Bhatnagar et al. 2003, Goldfrank 2007; Wampler 2012).
After the fall of dictatorship, democracy and citizens’ participation were the center
of attention (D). By the 1990s, more PBs were implemented in other countries of
South America outside Brazil, and the objectives of their realizations were similar.
An important role in the spreading of PBs in Europe was the World Social Forum,
an annual meeting of civic society organizations first held in Brazil in 2001, where
the concept of PB was presented (e.g. Sintomer et al. 2010). Many local govern-
ments from Europe found inspiration in PBs, and the first PBs were introduced
mostly in France, Spain and Italy and focused on SJ and D. The different aim of the
first generation of PBs implemented in Germany was to modernize public budget-
ing and to increase its efficiency (Ef) and to improve transparency (T) (Sintomer et
al. 2010, 2014; Baiocchi and Ganuza 2014; Ermert and Ruesch 2014; Ermert et al.
2015; Foelscher 2007). The first PB in a Visegrad country was realized in Poland in
2003 – 2005 (Džinić et al. 2016).
      Latter PBs were more focused on fostering the participation of citizens in
public issues and enhancing community development as a solution for the crisis
of democracy (D, C) (e.g. Foelscher 2007). These objectives could be achieved by
educating PB participants to be more active and informed citizens (Ed). Gondášová
and Murray Svidroňová (2016), for example, state that one of the effects of partici-
patory budgeting is increased citizen involvement when citizens play a crucial role
in the realization of the projects and in co-creation process in the type of PB Porto
Alegre adapted for Europe.

                                                                                113
Figure 2

114
                       Journey of PB from South America to Europe with the objectives for PB implementation

                                1990s                Early 2000s                  2003–2005
       Late 1980s              More PBs                                                                         Late 2010s
       First PBs in                               More PBs in Europe              First PB in
                               in South                                             Poland                   Next generation
          Brazil               America            France, Spain, Italy                                       of PBs in Europe
          SJ, D                                         SJ, D                      T, D, Ef
                                 SJ, D                                                                           C, Ed, D

                         1989                2001                        More PBs in
                      PB in Porto         Start of the                    Germany           More PBs in other CEE
                        Alegre            World Social                                            countries
                                            Forum                            Ef                   T, D, Ef
                        SJ, D

      Source: the authors
                                                                                                                                The NISPAcee Journal of Public Administration and Policy, Vol. XII, No. 2, Winter 2019 /2020
Does Participatory Budgeting Bolster Voter Turnout in Elections ? The Case of the…

1.1 Direct democracy and electoral turnout
Empirical studies provide ambiguous results about the impact of direct democracy
measures on voter turnout. Freitag and Stadelmann-Steffen (2010) distinguish two
approaches in assessing this relationship. First, prevailing empirical studies indicate a
positive relationship between direct democracy and voters’ participation, arguing that
direct democracy processes are associated with an increase in voter interest, political
information and knowledge (e.g. Tolbert and Smith 2005; Lassen 2005; Tolbert and
McNeal 2003; Smith 2001; Smith 2002; Dvořák et al. 2017). These results are in line
with the participatory theories of democracy (Barber 1984; Pateman 1970).
      The second approach insists on the opposite relationship – more direct de-
mocracy initiatives cause decreased significance of elections due to voter fatigue or
less perceived importance of direct electoral choices. Freitag and Stadelmann-Stef-
fen (2010) showed that the frequent use of direct democracy procedures resulted
in lower voter turnout in municipal and national elections as the result of less in-
terested potential voters. The theory of protest voting could be another explanation
for the negative impact of civic initiatives on the participation rate in elections. For
instance, Horton and Thompson (1962) and Buehlmann et al. (2003) consider local
referendums as possible institutional outlets for protests, leading to negative voting.
      The type of election also plays an important role in the evaluation of the im-
pact of direct democracy on voter turnout. Tolbert and Smith (2005) identified
different effects of citizen initiatives on voter turnout in presidential and midterm
elections in the USA. Turnout in presidential elections increases by 0.70 % with
each initiative on the ballot, with turnout in midterm elections increasing by 1.7 %.
Schlozman and Yohai (2008) identified increased turnout in midterm elections,
but no effect on turnout at presidential elections. Dvořák et al. (2017) distinguish
between the long- and short-term effects of direct democracy on voter turnout.
The education of citizens through the direct experience may increase their political
engagement in the long run. A public campaign connected with civic initiatives
could cause higher public interest but is rather election-specific and short-term.
Local direct democracy events have a positive impact on voter turnout, causing its
increase by 2.5 percentage points in local elections and by 1.5 percentage points in
parliamentary elections.
      Like direct democracy, the introduction of participatory budgets with the ob-
jectives to educate citizens, improve democracy and enhance community develop-
ment may influence voter participation in elections. Based on this assumption, we
formulated the following hypothesis:
     H: The introduction of participatory budgeting has a bigger positive effect on
voter turnout in municipal elections than on voter turnout in national elections.
      We presume, that voter turnout in municipal elections is more influenced by
the introduction of PB than is voter turnout in national elections, since participa-

                                                                                    115
The NISPAcee Journal of Public Administration and Policy, Vol. XII, No. 2, Winter 2019 /2020

tory budgets are focused on local issues and thus are more likely to change voters’
attitudes towards local politics.

2. Methodological framework
We tested the hypothesis by applying regression discontinuity analysis (OLS mod-
els). We used quasi-experimental analysis in line with the counterfactual approach
comparing the treatment group (municipalities with PB implementation) and
the control group (municipalities without PB). Quasi-experimental studies (also
known as natural experiments) are used to estimate a causal relationship and rely
on circumstances outside of the researcher’s control that naturally lead to random
assignment (Rosen and Gayer 2014).
      Characteristics of all compared subjects in both groups should be very sim-
ilar or the same on average (e.g. Potluka and Brůha 2013); therefore we used the
nearest-neighbor search method. We assume that by using this method we could
eliminate the probability of the existence of exogenous events that might influence
only one of the groups (Card and Krueger 1994). The application of the method of
searching for the nearest neighbors was based on the following main criteria: pop-
ulation (number of inhabitants); population density; location in the same NUTS III
as the municipality with PB.
      The population criteria can reduce the impact of the size of municipalities.
For example, Čmejrek (2007) states that small Czech municipalities have higher
voter turnout in municipal elections. The preferred location in the same NUTS III
is included to eliminate the impact of voting traditions or preferences typical for a
certain area. The criterion of the same NUTS III location was not used for Prague
districts – all municipalities in the control group were the Prague districts only
to eliminate the possible difference. We used a different approach for the city of
Ostrava because we could not find a proper pair based on the population number
or density in Ostrava districts only. We looked for similar municipalities in the same
NUTS III with a preference for the population criteria. Table 1 shows the different
sizes of municipalities in our experimental group.

                                        Table 1
                   Size of municipalities in the experimental group

                       Municipality size                    N            %

            Less than 10,000 inhabitants                    7            29

            10,000 to 100,000 inhabitants                   13           54

            More than 100,000 inhabitants                   4            17

           Source: the authors

116
Does Participatory Budgeting Bolster Voter Turnout in Elections ? The Case of the…

      The next step was to test for similarities in complementary criteria – expendi-
tures and revenues per one inhabitant. Expenditures per one inhabitant reflect the
extent of public services in the municipality. Revenues per one inhabitant are an
expression of potential financial sources that could be used for participatory bud-
geting. Both criteria are suitable for determining the relevance of PB application.
For the Ostrava and Prague districts the complementary criteria were not applied
due to the dissimilar basis and principles of the budgets of city districts compared
to the towns and cities in the control group.
     As the results of OLS models were not significant for all municipalities, we
applied also the method difference in differences used by Card and Krueger (1994).
This method assumes that the impact of the policy on the outcome can be calculat-
ed by means of two differences. The first difference is determined in time (before
and after intervention) and the second is caused by the difference between support-
ed and unsupported subjects (Potluka and Špaček 2013).
      We used public, web-based information about PBs in Czech municipalities,
and our own data was collected manually. For analyzing voter turnout, we used data
published by the Czech Statistical Office (CSO) on voter turnout in municipal and
parliamentary elections and population (density) for a certain municipality. Linked
open data on municipal expenditures and revenues from MONITOR (portal of the
Ministry of Finance of the Czech Republic) was used for testing the determination
of the control group. Our data set contains 65 PB cases in 30 municipalities in the
Czech Republic in the period 2016 – 2018. The substantial increase of PB cases start-
ed in the analyzed period, and their evolution is shown in Table 2.

                                         Table 2
                      Evolution of PB cases in Czech municipalities

                                              Average rate of     Amount used for
                           # of inhabitants
 Year       # of PBs                        total expenditures    all PB projects in
                               with PB
                                             used for PB in %            EUR

2016            13              556,510             0.66              1,407,000

2017            27             1,307,001            0.56              2,924,000

2018            25             1,402,216            0.59              3,702,000

Source: the authors

      Five municipalities did not meet the definitional criteria of PB based on Sin-
tomer et al. (2010) or the condition of at least 2 years duration of the PB process
in a certain municipality. Similarly, we had to exclude Brno because we could not
find a proper pair for the control group. After selection, there are 48 municipalities
in both the experimental and control groups used for OLS regression. We did two

                                                                                  117
The NISPAcee Journal of Public Administration and Policy, Vol. XII, No. 2, Winter 2019 /2020

OLS models – for all analyzed municipalities (N=48) and for the Prague districts
only (N=18).

                                           Table 3
               Descriptive statistics for the experimental group (N=24) 3 4

                                Mean             Std. Dev.           Min                Max

population                      42,612             38,120            1,133            109,790

expenditures (in EUR)        24,300,015         19,860,795         669,787          73,135,000

pb expenditures3               105,900            113,919            9,671            386,847

pb rate  4
                                 0.61               0.50             0.06               2.22

Source: the authors

2.1 Dependent and independent variables
Voter turnout in municipal elections and turnout in parliamentary (national) elec-
tions are considered dependent variables. We used voters’ participation rate for a
certain municipality in the control and the experimental groups in the last munici-
pal elections (2017) and in the last parliamentary elections (2018). The PBs were in-
troduced in this period, therefore we can test the immediate effect on voter turnout
after PB implementation.
      The independent variables were previous voter turnouts (in municipal and
parliamentary election), and the dummy variable was the presence of the PB. We
eliminated the differences in previous voter turnouts caused by possible unobserved
factors by using the nearest-neighbor search method. Previous voter turnouts (na-
tional and local) are similar for both groups (experimental and control), as shown
in Figure 3. Voter turnout in the period before PB implementation (2013 / 2014) is
on average the same for local and national elections in the control and experimental
groups. Voter turnout in the local election was less than voter turnout in the parlia-
mentary election during the whole analyzed period (by approximately 20 pp).
      We used municipal-level data about voter turnout in previous elections, both
in local and national elections (2013 and 2014), as dependent variables. We antici-
pated that there would be a positive correlation between the last and current voter
turnout. The presence of the PB was a dummy variable with value 1 in case of PB
application and 0 without PB. All municipalities in the experimental group have the
value 1, and the municipalities in the control group have the value 0.

3   pb expenditures are total expenditures in a municipality per year used for PB projects in EUR.
4   pb rate is the % of total expenditures used for PB projects.

118
Does Participatory Budgeting Bolster Voter Turnout in Elections ? The Case of the…

                                    Figure 3
      Voter turnout in the experimental and control group in % (2006 – 2018)

       70,00
       65,00
       60,00
       55,00
       50,00
       45,00
       40,00
                   2006          2010      2013/2014 2017/2018
                     loc_turnout_eperiment          loc_turnout_control
                     national_turnout_eperiment           national_turnout_control

    Source: developed by the authors, based on data from CSO (2019)

      When applying the difference in differences (DID) method, we compared the
average voter turnout in the municipalities in the control group (CG) with the ex-
perimental group (EG) in the previous and current elections. The method was used
for both municipal and parliamentary elections for all municipalities. Participatory
budgeting was introduced in the period after the previous elections, therefore we
can evaluate its impact on voter turnout.

3. Results and discussion
Table 4 presents the results of the OLS regression models for all municipalities. Our
results are in line with the expectation that voter turnout in the previous elections
be positively correlated with the current voter participation in both national and
local elections. However, we did not find evidence that the introduction of PB had a
significant impact on voter turnout in all municipalities.
      When we applied our model to the data set for Prague districts only (Table
5), we found a positive correlation for PB use and voter turnout in both local and
national elections. Nevertheless, there is a difference in the level of significance and
coefficient. Voter turnout in municipal elections is more significant than voter turn-
out in parliamentary elections.

                                                                                     119
The NISPAcee Journal of Public Administration and Policy, Vol. XII, No. 2, Winter 2019 /2020

                                     ***Table 456789
       Empirical results of the regression analysis (the OLS models) with N=48

            Dependent variable munelect_15                R=0.652371, R2=0.636921

                              Standard                                    Hypothesis    Fact
                Coefficient                t-stat      p-value
                              error                                       (correlation) (correlation)
Const           11.6769       3.9128       2.9843      0.00458***

pb              1.51629       1.69483      0.8947      0.37573            ambiguous       insignificant

munelect_06 0.836978          0.0917581 9.1216
Does Participatory Budgeting Bolster Voter Turnout in Elections ? The Case of the…

      The implementation of PB increased voter turnout in municipal elections by
approximately 3 percentage points and by approximately 1.2 percentage points in
parliamentary elections. We presumed that the positive effect of PB use on voter
turnout was bigger in the case of municipal elections than in the case of parliamen-
tary elections, and the hypothesis was confirmed.
     Consequently, we analyze differences in voter turnout for all municipalities
before and after the PB implementation using the difference in differences (DID)
method. Table 6 presents the results for municipal elections.

                                    Table 6
 Municipal election – DID in voter turnout in all municipalities before and after
                         the PB implementation in %

                              Experimental       Control group      difference
                               group (EG)            (CG)            (EG-CG)
      munelect_0                   41.04             40.59             0.45

      munelect_1                   47.54             45.65             1.89

      difference in time           6.50               5.06             1.44

     Source: the authors

      In both groups, there was an almost identical increase in the average voter
turnout by over 5 percentage points in time. In the first period (before PB introduc-
tion), the difference between the experimental and control groups was less than 0.5
percentage points. The relative gain (the difference in differences) of the changes in
voter turnout was 1.44 %.

                                      Table 7
  Parliamentary election – DID in voter turnout in all municipalities before and
                        after the PB implementation in %

                              Experimental       Control group      difference
                               group (EG)            (CG)            (EG-CG)
      parelect_0                   62.02             61.31             0.71

      parelect_1                   64.36             63.42             0.94

      difference in time           2.35               2.11             0.23

     Source: the authors

      Table 7 shows that on average the increase in voter turnout was by approxi-
mately 2 percentage points in time, which is less than the result in municipal elec-
tions. Similarly, there was only a weak impact of PB implementation on the average
voter turnout in the experimental group with a relative gain of 0.23 %. As shown in

                                                                                 121
The NISPAcee Journal of Public Administration and Policy, Vol. XII, No. 2, Winter 2019 /2020

Tables 6 and 7, the positive effect of PB use on voter turnout was present in both
cases, but there must be an emphasis on the difference in the type of election. The
DID in voter turnout in all municipalities before and after the PB implementation
was bigger by 1.21 percentage points in the case of municipal elections than in the
case of parliamentary elections.
      The results of the OLS models for the Prague districts and the difference in
differences approach show that PB implementation increased voter turnout in
municipal elections more than in parliamentary elections. The results of the OLS
models for all municipalities suggest that voter turnout in both local and national
elections was almost unaffected by the PB introduction; they were statistically in-
significant. We tried to eliminate other factors influencing voter turnout by using
the nearest-neighbor method to select a suitable pair of municipalities, but we are
aware of its imperfections. Different results for the dataset with all municipalities in
comparison with the Prague districts could be caused by other socio-demograph-
ic and economic factors that could influence the results significantly and must be
tested in further research. The OLS regression model could be amended by adding
more independent variables reflecting this absence. The limitation of the model is
the lower number of observation due to the short period of PB application. On the
other hand, we can exclude Horton and Thompson’s (1962) theory about negative
voting of, because there is no negative correlation between PB introduction and
voter turnout on average for all municipalities.
      Dvořák et al. (2017) mentioned another confounding factor influencing voter
turnout. This could be a conflict within a concrete municipality that might energize
the local community. We did not have information on all circumstances in the an-
alyzed municipalities where that could be the case. Therefore, another option for
further research could be more qualitative analysis, e.g. in the form of case studies
about selected municipalities focused on the local political situation and about a
possible conflict in the community that could occur before elections.
       Additionally, Džinić (2018) states that the significance of the local population’s
participation in managing public finances through PB depends on the degree of
fiscal decentralization and whether the citizens manage a substantial share of public
finance. In our experimental group the mean PB rate (% of total expenditures used
for PB projects) was 0.61, and the maximum was 2.22. In the CEE region, the situa-
tion is similar in Poland, where the PB rate in 2018 was below 1 % with a maximum
of 0.85 in Katowice (Olejniczak and Bednarska-Olejniczak 2018). It could be ques-
tionable to consider this rate to be high enough to influence civic participation and
motivate individuals to vote.
     The size of a municipality can be an important factor that can influence PB
application. A small municipality often implies limited financial resources, which
may make a more large-scale implementation of PB more complicated (Krenjova
and Raudla 2013). In the CEE region there are highly fragmented territorial struc-

122
Does Participatory Budgeting Bolster Voter Turnout in Elections ? The Case of the…

tures of local government, and they tend to be rather small (Swianiewicz 2010;
Illner 1998), which is typical of the Czech Republic with more than 6 thousand
municipalities.
      Furthermore, the real participation of citizens in the PB process can play an
important role in the evaluation of PB’s influence on voter turnout. Zepic et al.
(2017) identified some reasons why inhabitants refuse to participate in the PB pro-
cess, such as a relatively small PB budget, the low trust in the implementation of PB
projects etc. In some municipalities in Poland voter turnout in votes on the partic-
ipatory budget was more than 30 % in 2016 (Olejniczak and Bednarska-Olejniczak
2018). However, based on our preliminary data voter turnout on PB in Czech mu-
nicipalities was at a maximum of around 16 % in smaller municipalities for the
whole analyzed period, and it differs a lot. In the municipalities with weak interest
in PB participation and PB voting, it is arguable to expect any exact causal mech-
anism of how PB introduction increases voter turnout in local / national elections.
       The results of OLS regression for the Prague districts are in line with Dvořák
et al. (2017) testing the impact of the use of municipal referendums on voter turn-
out. There was an increase of voter turnout by 2.5 pp in municipal election and 1.5
pp in parliamentary elections, which is close to our results (3 percentage points in
municipal election and 1.2 percentage points in parliamentary election). Likewise,
the higher impact of PB implementation on voter turnout in local elections than in
national elections was shown when we applied the difference in differences method.
The difference in differences of the changes in voter turnout before and after PB
implementation is 1.44 % in the case of municipal elections, which was more than
the relative gain of 0.23 % in parliamentary elections. In addition, our findings are
compatible with Tolbert and Smith (2005), who found evidence for the different
impact of citizen initiatives on voter turnout in presidential and midterm elections.
The effect on the midterm elections was higher by 1 pp than on the presidential
election, which could be caused by the higher concentration on local issues akin to
our results concerning the Czech local elections.
     We analyzed the impact of PB introduction on voter turnout on the aggregat-
ed municipal level. Another option could be the analysis on the individual level,
which was applied in Johnson et al. (2018). Engaging with PB caused the increased
individuals’ probability of voting by an average of 7.5 %, with a greater effect on the
groups of citizens who have lower probabilities of voting (e.g. young voters). Un-
fortunately, nowadays there is no data about individuals voting in PB available for
Czech municipalities.

Conclusion
The objective of the paper was to evaluate whether the use of participatory budgets
is associated with a higher voter turnout in parliamentary and municipal elections.

                                                                                  123
The NISPAcee Journal of Public Administration and Policy, Vol. XII, No. 2, Winter 2019 /2020

We provide some support for a positive and significant effect of PB implementation
in the Prague districts in both local and national elections. However, the results for
all municipalities in our dataset were not significant. When reflecting the types of
elections, we found that the impact of PB use on voter turnout in local elections is
higher than in national elections (for all municipalities), which is in line with our
assumptions.
     Thus, participatory budgeting could increase voter turnout in local elections,
but there are other factors that must be considered, and there is a need to analyze
these factors more deeply. Solving the “crisis of democracy” by means of PB im-
plementation alone is not probable, and local politicians should be rather prudent
and not consider PB as an ideal instrument to foster civic participation under all
circumstances.
      Even if PB is often considered an innovative tool to promote the concept of
good governance, based on the results of our study it is not possible to confirm its
impact on a higher participation rate and voter turnout without a deeper analysis
of other factors that could be specific for a concrete municipality. Therefore, we rec-
ommend a more qualitative approach in researching this area and in decision-mak-
ing when introducing PB. A good example of providing this type of research on a
particular municipality could be the case study (Špaček 2018) of Brno’s 2050 Strat-
egy with a view to enhancing inclusivity and participation.
       Additionally, the aim of the study was to create a unique dataset about proj-
ect-oriented PBs in the Czech Republic using public web-based information with
the intention to have better insight into the development of Czech PB cases. Further
research could be a systematic overview of Czech PB cases focused on other poten-
tially problematic issues than voter turnout and civic participation.
      The period of PB application in the Czech Republic is rather short, and there-
fore adding more finished PB cases in Czech municipalities to our dataset in the
future might show results with a potentially higher explanatory power.

Acknowledgment
We would like to thank Jakub Šácha for discussions and suggestions on the empiri-
cal framework of the paper, and we are grateful to Troy Mix for comments.

References
Baiocchi, Gianpaolo and Ernesto Ganuza. 2014. “Participatory Budgeting as if
      Emancipation Mattered.” Politics & Society 42(1), 29 – 50.
Barber, R. Benjamin. 1984. Strong Democracy: Participatory Politics for a New Age.
      Berkeley: University of California Press.

124
Does Participatory Budgeting Bolster Voter Turnout in Elections ? The Case of the…

Berman, M. Evan. 1997. “Dealing with Cynical Citizens.” Public Administration Re-
     view 57(2), 105 – 112.
Bhatnagar, Deepti, Animesh Rathore, Magui Moreno Torres and P. Kanugo. 2003.
     Participatory Budgeting in Brazil (English). Empowerment case studies.
     Washington. DC: World Bank.
Boc, Emil. The Development of Participatory Budgeting Processes in Cluj-Napoca,
      Transylvanian Review of Administrative Sciences, 58 E/2019, 38-51.
Buehlman, Marc, Markus Freitag and Adrian Vatter. 2003. “Die schweigende Meh-
     rheit: Eine Typologie der Schweizer Nichtwählerschaft.” In: P. Sciarini, S.
     Hardmeier and A. Vatter (eds). Schweizer Wahlen 1999. Swiss Electoral Stud-
     ies (5). Bern: Paul Haupt, 27 – 58.
Cabannes, Yves. 2004. “Participatory Budgeting: A Significant Contribution to Par-
     ticipatory Democracy.” Environment and Urbanization 16(1), 27 – 46.
Card, David and B. Krueger Alan. 1994. “Minimum Wages and Employment: A
      Case Study of the Fast-Food Industry in New Jersey and Pennsylvania.”
      American Economic Review 84(4), 772 – 793.
Carroll, Cecile, Thea Crum, Carolina Gaete, Maria Hadden and Rachel Weber. 2016.
      Democratizing Tax Increment Financing Funds through Participatory Budget-
      ing. Chicago: University of Illionis, 67 pgs. Available at https://irrpp.uic.edu/
      pdf/publications/TIF-PB-Toolkit-June-2016.pdf (last accessed 8 July 2019).
Czech Statistical Office (CSO). 2019. “Výsledky voleb a referend” [Results of elec-
     tions and referendums]. Available at https://volby.cz/ (last accessed 17 July
     2019).
Čmejrek, Jaroslav. 2007. “Citizens’ Local Political Participation in the Czech Repub-
     lic: Rural-Urban Comparison.” Agricultural Economics (AGRICECON) 53(1),
     21 – 29.
Dvořák, Tomáš, Jan Zouhar and Jakub Novák. 2017. “The Effect of Direct Democ-
     racy on Turnout: Voter Mobilization or Participatory Momentum ?” Political
     Research Quarterly 70(2), 433 – 448.
Džinić, Jasmina. 2018. “Participatory Budgeting in Croatia: A Way to Improve
      Quality of Local Governance ?”. In Juraj Nemec et al. (eds). Alternative Deliv-
      ery Service. Brussels: IASIA / IIAS. ISBN 978-2-931003-01-5, 73 – 81.
Džinić, Jasmina, Mária Murray Svidroňová and Ewa Markowska_Bzducha. 2016.
      “Participatory Budgeting: A Comparative Study of Croatia, Poland and Slo-
      vakia.” NISPAcee Journal of Public Administration and Policy 9(1), 31 – 56.

                                                                                  125
The NISPAcee Journal of Public Administration and Policy, Vol. XII, No. 2, Winter 2019 /2020

Ermert, Julian, Hannah Pützer, Michelle Ruesch, Zebralog GmbH & Co KG.
     2015. 8. Statusbericht des Portals Buergerhaushalt.org. Available at https://
     www.buergerhaushalt.org/sites/default/files/downloads/8._Statusbericht_
     Buergerhaushalte_in_Deutschland_Juni_2015.pdf (last accessed 8 July 2019).
Ermert, Julian and Michelle Ruesch. 2014. 7. Statusbericht des Portals Buerger-
     haushalt.org. Available at https://www.buergerhaushalt.org/sites/default/
     files/7._Statusbericht_2014_Buergerhaushalte_in_Deutschland_13062014.
     pdf (last accessed 8 July 2019).
European Parliament. 2014. Results of the 2014 European elections. Available at
     http://www.europarl.europa.eu/elections2014-results/en/turnout.html (last
     accessed 8 July 2019).
European Parliament. 2019. Results of the 2019 European elections. Available at
     https://www.election-results.eu/turnout/ (last accessed 8 July 2019).
Foelscher, Alta. 2007. “Participatory Budgeting in Central and Eastern Europe.” In
      A. Shah (ed.). Participatory Budgeting. World Bank Publications, Public Sec-
      tor Governance and Accountability Series. Washington D.C.: The Interna-
      tional Bank for Reconstruction and Development / The World Bank, 127 – 156.
Freitag, Markus and Isabelle Stadelmann-Steffen. 2010. “Stumbling Block or Step-
       ping Stone ? The Influence of Direct Democracy on Individual Participation
       in Parliamentary Elections.” Electoral Studies 29(3), 472 – 483.
Goldfrank, Benjamin 2007. “Lessons from Latin American Experience in Partic-
      ipatory Budgeting.” Participatory Budgeting. Presented at the Latin Amer-
      ican Studies Association Meeting San Juan. Puerto Rico. March 2006. 143,
      91 – 126.
Gondášová, Lýdia and Mária Murray Svidroňová. 2016. “Participatory Budget-
     ing as an Innovation in Local Public Service Delivery: The Slovak Case.” In
     Špalková, Dagmar and Matějová Lenka (ed.). Current Trends in Public Sector
     Research. Brno: Masaryk University, 248 – 256.
Haken, Roman, František Havlin, Ondřej Marek, Pavel Mička, Martin Nawrath,
     Zuzana Vachůnova, Aleš Ziegler. 2015. Analyza občanske participace v ČR.
     Prague: Agora CE.
Horton, E. John and Wayne E. Thompson. 1962. “Powerlessness and Political Nega-
     tivism: A Study of Defeated Local Referendums.” American Journal of Sociol-
     ogy 67(5), 485 – 493.
Illner, Michal. 1998. “Territorial Decentralization: An Obstacle to Democratic Re-
        form in Central and Eastern Europe ?” In Jonathan D. Kimball (ed.). The
        Transfer of Power: Decentralization in Central and Eastern Europe. Budapest:
        Local Government and Public Service Reform Initiative, 7 – 42.

126
Does Participatory Budgeting Bolster Voter Turnout in Elections ? The Case of the…

Jacobi, Pedro. 1999. “Challenging Traditional Participation in Brazil: The Goals of
       Participatory Budgeting.” Washington DC: Woodrow Wilson International
       Center for Scholars. Comparative Urban Studies Occasional Papers Series 32.
Johnson, Carolina, Jacob H. Carlson and Sonya Reynolds. 2018. “Participatory De-
      mocracy and Voter Turnout: Evidence from Participatory Budgeting.” Work-
      ing Paper. Available at https://www.hjacobcarlson.com/files/PBTurnout2018.
      pdf (last accessed 8 July 2019).
Krenjova, Jelizaveta and Ringa Raudla. 2013. “Participatory Budgeting at the Local
      Level: Challenges and Opportunities for New Democracies.” Halduskultuur –
      Administrative Culture 14(1), 18 – 46.
Lassen, D. David. 2005. “The Effect of Information on Voter Turnout: Evidence from
      a Natural Experiment.” American Journal of Political Science 49(1), 103 – 118.
Matovu, George and Takawira Mumvuma. 2008. Municipal Finance Programme
     Part II Participatory Budgeting. Manual. Municipal Development Partnership
     for Eastern and Southern Africa. Nairobi: United Nations Human Settlements
     Programme and Municipal Development Partnership for Eastern and South-
     ern Africa. ISBN 978-92-1-131954-5, 87 pgs.
Olejniczak, Jarosław and Dorota Bednarska-Olejniczak. 2018. “Participatory Bud-
      geting in Poland in 2013 – 2018: Six Years of Experiences and Directions of
      Changes.”, 337 – 354. In Dias, Nelson (ed.). Hope for Democracy: 30 Years of
      Participatory Budgeting. Epopeia Records. Available at http://npms.cfh.ufsc.
      br/files/2018/09/hope_for_democracy_-_digital.pdf (last accessed 8 July
      2019).
Pateman, Carole. 1970. Participation and Democratic Theory. Cambridge: Cam-
     bridge University Press.
Potluka, Oto and Jan Brůha. 2013. “Zkušenosti s kontrafaktuální dopadovou evalu-
      ací v České republice.” Evaluační teorie a praxe 1(1), 53 – 68.
Potluka, Oto and Martin Špaček. 2013. “Postupy a metody kontrafaktuálních dopa-
      dových evaluací pro Operační program Zaměstnanost v období 2014 – 2020:
      Ministerstvo práce a sociálních věci.” Available at https://www.mpsv.cz/files/
      clanky/17051/Metodika_CIE_MPSV_131015.pdf (last accessed 8 July 2019).
Rosen, S. Harvey and Ted Gayer. 2014. Public Finance. Singapore: McGraw Hill.
Schlozman, Daniel and Ian Yohai. 2008. “How Initiatives Don’t Always Make Citi-
      zens: Ballot Initiatives in the American States, 1978 – 2004.” Political Behavior
      30(4), 469 – 489.

                                                                                  127
The NISPAcee Journal of Public Administration and Policy, Vol. XII, No. 2, Winter 2019 /2020

Sintomer, Yves, Carsten Herzberg, Giovanni Allegretti and Anja Roecke. 2010.
      Learning from the South: Participatory Budgeting Worldwide – an Invitation
      to Global Cooperation. Study No. 25. Dialog Global series. Bonn: InWEnt
      gGmbH – Capacity Building International.
Sintomer, Yves, Carsten Herzberg, Giovanni Allegretti and Anja Roecke. 2013.
      Participatory Budgeting Worldwide. Updated Version. Study No. 25. Dialog
      Global series. Bonn: Engagement Global.
Sintomer, Yves, Carsten Herzberg, Giovanni Allegretti and Anja Roecke. 2014.
      Bürgerhaushalte weltweit. Aktualisierte Studie No. 25. Bonn: Engagement
      Global.
Sintomer, Yves, Carsten Herzberg and Anja Roecke. 2008. “Participatory Budgeting
      in Europe: Potentials and Challenges.” International Journal of Urban and Re-
      gional Research 32(1), 164 – 178. Available at http://onlinelibrary.wiley.com/
      doi/10.1111/j.1468-2427.2008.00777.x/full (last accessed 17 July 2019).
Smith, A. Daniel. 2001. “Homeward bound ? Micro-Level Legislative Responsive-
      ness to Ballot Initiatives.” State Politics and Policy Quarterly 1(1), 50 – 61.
Smith, A. Mark 2002. “Ballot Initiatives and the Democratic Citizen.” Journal of
      Politics 64(3), 892 – 903.
Swianiewicz, Pavel. 2010. “If Territorial Fragmentation is a Problem, is Amalgama-
      tion a Solution ? An East European Perspective.” Local Government Studies
      36(2), 183 – 203.
Špaček, David. 2018. “Strategy Formulation in a Large City: The Case of Brno 2050
      Strategy.” Working Draft prepared for the IRSPM conference, 11 – 13 April
      2018, Edinburgh, UK.
Tolbert, J. Caroline, John A. Grummel and Daniel A. Smith. 2001. “The Effects of
      Ballot Initiatives on Voter Turnout in the United States.” American Politics
      Research 29(6), 625 – 648.
Tolbert, J. Caroline and Ramona S. McNeal. 2003. “Unraveling the Effects of the
      Internet on Political Participation ?” Political Research Quarterly 56(2),
      175 – 185.
Tolbert, J. Caroline and Daniel A. Smith. 2005. “The Educative Effects of Ballot Ini-
      tiatives on Voter Turnout.” American Politics Research 33(2), 283 – 309.
UN-Habitat. 2004. 72 Frequently Asked Questions about Participatory Budget-
    ing. HS / 722 / 04E. Available at https://unhabitat.org/books/72-frequent-
    ly-asked-questions-about-participatory-budgeting/ (last accessed 14 July
    2018).
Wampler, Brian. 2007. “Guide to Participatory Budgeting.” In A. Shah (ed.). Partici-
    patory Budgeting. Washington D.C.: World Bank Publications, 21 – 53.

128
Does Participatory Budgeting Bolster Voter Turnout in Elections ? The Case of the…

Wampler, Brian. 2012. “Participatory Budgeting: Core Principles and Key Impacts.”
    Journal of Public Deliberation 8(2): Article 12.
Zepic, Robert Marcus Dapp and Helmut Krcmar. 2017. “Das Geheimnis geringer
      Beteiligungsquoten deutscher Bürgerhaushalte.” Springer Fachmedien Wies-
      baden 2017. Available at http://www.buergerhaushalt.de/de/article/publika-
      tion-erschienen-e-partizipation-und-keiner-macht-mit (last accessed 8 July
      2019).

                                                                               129
You can also read